local intelligence

can it run

Can the AMD Radeon Instinct MI60 run gemma3 4B?

FULL

3.8 GB needed of 32 GB usable — headroom for context

32GB · GCN5.1 · model file 3.3GB · 4B

~87–140 tok/s

computed roofline: 1024 GB/s × 0.28–0.45 efficiency window / 3.3GB (Q4_K_M) — bands, never points. Small models (under ~4.5B) run BELOW these bands — kernel-launch overhead keeps them from saturating bandwidth (measured against community medians, 09-06).

Run gemma3 4B on instead

Raspberry Pi 5 + AI Kit (Hailo-8L) — cheapest catalog machine that runs gemma3 4B fully ($110)

RTX 3090 (24GB, used) — upgrade path machine

every model the AMD Radeon Instinct MI60 can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the AMD Radeon Instinct MI60 · fit = weights + 4k KV vs usable memory